Experimental Core ?Project Summary The goal of the Experimental Core is to be able to generate from thousands of chemicals a spectrum of their possible metabolites and simultaneously identify the metabolites and the enzyme(s) responsible. This will be achieved via development of 96-well, then 384-well and eventually 1536-well arrays of cells expressing individual xenobiotic metabolizing enzymes to generate the metabolites, which will be analyzed by high- resolution mass spectrometry. Enzymes will be expressed in human Huh7 hepatoma cells using lentiviral transduction, which we have used successfully to express several xenobiotic-metabolizing cytochrome P450 (P450) enzymes in catalytically active form. We will begin by expressing major enzymes in the P450, flavin monooxygenase, UDP-glucuronyltransferase, sulfotransferase and N-acetyltransferase families and continue to develop additional cell lines through the duration of the project. Arrays will be plated and screened using robotic facilities in the Miller laboratory and the Emory Chemical Biology Discovery Center. The Computational Core will generate a list of candidate xenobiotic metabolic products, allowing us to determine whether a product with accurate mass matching the predicted metabolite is formed. Using the predicted m/z of the predicted metabolite, we will be able to direct MS/MS acquisition for likely products and compare those to MS/MS predicted for the metabolite by the Computational Core, thereby providing two levels of chemical identification in an automated manner. Broad coverage of tissue metabolites will be obtained by hybrid arrays incorporating pooled S9 fractions from human tissues together with the cellular approach. Since we will have a cell line specifically expressing an enzyme that generates a predicted product, we will have a scalable system for biologically producing enough of the xenobiotic metabolite to perform more definitive chemical identification. The first-generation platform will utilize pooled human S9 tissue fractions in 96-well format to rapidly develop our capacity to generate metabolites of important environmental chemicals. The second-generation platform will be a 96-well array of cell lines developed in the first year. The third-generation platform will be in 384 well format with an expanded array of enzymes. All platforms will be tested against a panel of model environmental chemicals with known metabolites. In the fourth year, the 3rd generation platform will be used to catalogue the metabolites of >200 chemicals from the EPA's TOXCAST library, and work towards building a 1536-well platform. The approach is scalable to support analysis of dozens of metabolites of multiple chemicals per day, providing a prospect for mega-scale chemical identification.